Public Sector Accounting Discussion Group

Report on the Public Meeting

October 27, 2015
The Public Sector Accounting (PSA) Discussion Group is a discussion forum only. The Group’s
purpose is to support the Public Sector Accounting Board (PSAB) by enabling discussion in a public
venue of issues arising from the application of the CPA Canada Public Sector Accounting Handbook
(PSA Handbook). The Group comprises members with various backgrounds who participate as
individuals in the discussion. Any views expressed in the public meeting do not necessarily represent
the views of the organization to which a member belongs or the views of PSAB. The discussions of the
Group do not constitute official pronouncements or authoritative guidance.

This document has been prepared by the staff of PSAB and is based on discussions during the Group’s
meeting.

Comments made in relation to the application of the PSA Handbook do not purport to be conclusions
about acceptable or unacceptable application of the PSA Handbook. Only PSAB can make such a
determination.

Items Presented and Discussed at the October 27, 2015 Meeting

Sections PS 3200, PS 3250 and PS 3255: Vacation Paid in the Year of Retirement

Section PS 3150: Acquisition and Development of Software Associated with Cloud Computing

Sections PS 3260: Liability for Contaminated Sites

Standard Setting in the Public Interest

 Report on Public Meeting on October 27, 2015 – Non-authoritative material

ITEMS PRESENTED AND DISCUSSED

Sections PS 3200, PS 3250 and PS 3255: Vacation Paid in the Year of
Retirement
Scenario

In the year of retirement employees receive their full annual vacation entitlement regardless of when
nd
they retire in the calendar year. As a result, the vast majority of employees retire on January 2 of each
nd
year. By working to January 2 , the employee will receive five to seven weeks of vacation time (the
typical vacation allotment for senior employees) paid out upon retirement. To receive this vacation time
pay out on retirement, individuals must meet retirement criteria, which is typically an age plus years of
service criteria. There are individuals that retire at other points of the year but the vast majority will
defer retirement to obtain this additional benefit.

In both issues submitted, the question was first whether there is a liability as defined in Section PS
3200, Liabilities, for vacation pay to be accrued or whether it was a period cost. If a liability is to be
accrued, the second question is then whether such accrual would be based on the vesting and
accumulating provisions for liability recognition in Section PS 3250, Retirement Benefits, or the event-
driven provisions in Section PS 3255, Post-employment Benefits, Compensated Absences and
Termination Benefits.

Issue – Is the vacation paid in the year of retirement within the scope of Section PS
3255?
Group members sought clarification and it was determined that variations on the submitted scenario
exist across the country, leading to a conclusion that different fact scenarios may lead to different
answers to the question. It was decided to contrast the accounting for vacation entitlements by retirees
with that done for all other employees in order to focus the discussion on whether the vacation
entitlement benefit in the year of retirement was earned at the beginning of the year in which retirement
occurs or over the years to retirement.

The Group’s Discussion

Three views were considered:

View A – Vacation pay is a short-term employee benefit and not within the scope of Section PS 3255

View B – The vacation paid in the year of retirement is within the scope of Section PS 3255 as it is a
post-employment benefit provided by the government as a result of an event or events occurring

View C – The vacation paid in the year of retirement is not within the scope of Section PS 3255 but
rather within the scope of Section PS 3250 as it is a type of post-employment benefit that vests or
accumulates over the periods the employee provides services

The Group’s discussion focused primarily on whether recognition should be event driven (i.e., the
nd
employee’s choice of a retirement date of January 2 , or if eligibility criteria, such as age and years of
service should also affect recognition. In the course of the discussion, further facts were set out in
order to test whether comparable situations would give rise to consistent accounting treatments:

2
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

 A query was raised as to whether just being employed at the beginning of the year is sufficient to
earn the annual vacation entitlement; for all employees, not just those retiring in the year.

 A further query was whether a different approach would be taken in accounting for a scenario in
which a retiree took his vacation entitlement rather than being paid out and who therefore retired
after the vacation entitlement was used (i.e., a mid to late February retirement date rather than
nd
January 2 ). Only the timing of the payment would differ.

 And it was asked whether employees that leave, but do not retire, during the year would also be
entitled to vacation pay for the full year entitlement regardless of when they left as long as they
were employed at the beginning of the year.

Group members generally concluded that vacation earned by employees not retiring, even if considered
to be earned at the first of the year, would likely be an expense of the period, similar to salary expense.
And many members also agreed that if no other criteria than employment at the first of the year was
nd
required for retirees with a retirement date of January 2 to earn their vacation entitlement for the year,
then the cost of their vacation entitlement for the year of retirement may also be a period expense in the
year of retirement.

However, others felt that the vacation entitlement for a retiree in the year of retirement was actually
earned either as a result of meeting age and years of service eligibility criteria or earned more generally
over the course of the retiree’s employment. And there was some concern expressed about booking an
expense for the entire amount in the year of retirement rather than accruing a liability over the periods
retirees earn their final vacation entitlement through service. Questions were also raised regarding the
nature of the payout of such a retiree’s final vacation entitlement and whether it would be akin to event-
driven items such as a retirement allowance or certain kinds of severance payments.

There was some concern that the accounting for vacation entitlements for retirees in the year of
retirement should not be unnecessarily complex for amounts that were unlikely to be material and that
the cost of a more complex approach, such as accruing some kind of actuarially determined liability
amount for the employee group as a whole over time, might not be justified by the benefit of being
theoretically correct. And it was not clear whether the standards would actually demand such
complexity and related cost.

It was acknowledged that the fact set in the scenario did not indicate whether this benefit would vest at
a certain time in the employee’s service. The scenario also did not label the vacation entitlement as
solely a type of retirement benefit. The Group decided to focus its remarks on three possibilities:

 If an employee meets the retirement criteria and retires early in the fiscal year, the employee would
get the five to seven weeks of vacation entitlement paid on retirement.

 If all employees generally earn/accrue their whole annual vacation entitlement on the first of the
year, and if they leave the organization mid-year, they will still have earned the full entitlement by
being employed on the first of the year.

3
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

 If the vacation entitlement is earned by retirees on the first of the year but there is also a
requirement that they retire in that year, then the employee could choose to take the vacation and
retire after taking the five to seven weeks’ vacation.

There was some feeling that if the accounting treatment is event driven, then there may still be a public
interest requirement to accrue something in advance if there is knowledge that a lot of such events are
coming up; accountability to the public might require such an accrual. It was noted that certain kinds of
severance payments that can be anticipated would likely be accrued in accordance with Section PS
3255 and the scenarios being discussed could be considered comparable to such severance payments.
In contrast, another Group member noted that a cost/benefit evaluation and a conclusion not to accrue
might also be in the public interest.

The Group concluded its discussion by articulating the following:

 If there exists an eligibility criteria demanding that the individual retire, meet retirement age and
years of service requirements, as well as be employed at the first of the fiscal year, then View B is
likely to apply as this would be a fundamentally event-driven scenario. Accrual would occur as the
eligibility criteria are met, likely at the retirement date.

 However, there may be a public interest concern to consider if the amounts are significant and all
employees generally choose retirement dates a couple of days into a new fiscal year. In such
circumstances, there may be an argument for View C, which would require accrual of estimated
amounts to be paid out, likely calculated through actuarial estimate, over the periods services are
provided. In such circumstances, meeting the years of service criterion might be considered to be
met over the periods of service rather than when a designated number of years are reached.

 In contrast, if all employees earn the vacation entitlement on the first day of a new fiscal year, then
the amounts are merely employee benefits provided during the year and would be a period cost in
that year consistent with View A, even for the retirees.

Section PS 3150: Acquisition and development of software associated with

cloud computing
Software that is not stored on hardware owned or physically located on the entity’s premises and that
may be subject to a long-term licensing arrangement as opposed to ownership is known as “cloud
computing”.

Section PS 3150, Tangible Capital Assets, describes tangible capital assets as having physical
substance. However, footnote 2 in Section PS 3150 indicates that tangible capital assets include
computer software. At the time of writing the standard, it was argued that software was integral to the
use of computer hardware and should be accounted for in the same way as the computer hardware.

When the Section was released, there was little if any cloud computing, and virtually all software was
stored or the code written on hardware owned by, and itself stored on the premises of, the owner. In
addition, the vast majority of software was purchased outright and assumed to be owned by the
purchaser. Since that time, a shift in the way entities acquire and house software has occurred.

4
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

The presenter of the issue indicated that both the IASB and FASB were also looking at issues related to
accounting for cloud computing.

The National Institute of Standards and Technology, part of the U.S. Department of Commerce,
provides the following definition of cloud computing:

“Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a
shared pool of configurable computing resources (e.g., networks, servers, storage, applications,
and services) that can be rapidly provisioned and released with minimal management effort or
service provider interaction. This cloud model is composed of five essential characteristics: On-
demand self-service, Broad network access, Resource pooling, Rapid elasticity, Measured service;
three service models: Software as a Service (SaaS), Platform as a Service (PaaS), Infrastructure
as a Service (IaaS); and four deployment models: Private cloud, Community cloud; Public cloud,
and Hybrid cloud."

The Group’s Discussion

The Group spent some time discussing various aspects of cloud computing and software based in the
cloud in order to clarify its understanding before discussing the specific issues raised by submitters.

One nuance identified was that the servers where the cloud is hosted comprise the hardware; however,
personal computers or similar devices are needed in order to operate and access the software based in
the cloud.

Group members then spent some time clarifying terms and the exact fact sets for the scenarios to be
discussed. It was clarified that “acquired” generally means “purchased” and that the software is stored
in the cloud. However, it was noted that contracts vary regarding the software and whether it could be
taken off the cloud by an acquirer and used in house. Some software is customized for a single client
or group of clients while other software stored in the cloud is less customized and used by a multitude
of clients with similar needs.

PSAB staff clarified why intangibles were scoped out of Section PS 3150. One member indicated that
straightforward intangibles such as those held by private sector entities could be accommodated in
public sector financial statements. Staff indicated that the variety of public sector intangibles is broader
than in the private sector and includes unique sovereign and other rights, such as air and sea rights,
and rights to the electromagnetic spectrum. PSAB plans to look at all government intangibles together
in the future. However, for now, in both the conceptual framework and Section PS 1201, Financial
Statement Presentation, intangibles are not recognized in public sector financial statements:

“In the absence of appropriate public sector recognition and measurement criteria for intangibles, all
intangibles, including those that have been purchased, developed, constructed or inherited in right
of the Crown, are not recognized as assets in government financial statements. The intangible often
called "human capital" that embodies the talent or intellectual capital of government employees is
not recognized as an asset under existing accounting principles.”

5
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

Issue – Is the software associated with cloud computing an asset of the entity to be
recognized on the statement of financial position and expensed through amortization
over a period of “useful life” or should the costs be expensed as incurred?
Scenario 1

An entity has acquired software based in the cloud. The entity does not own the hardware associated
with the cloud-based software and may not know where that hardware physically exists. Rather, the
entity has contracted with a third party to host the software.

View A – The software is a tangible capital asset

In order for this view to be acceptable, one must conclude that the software meets the definition of an
asset and that the inclusion of software in the definition of a tangible capital asset extends to cloud-
based software. This view assumes that a bifurcation approach to hardware and software is permitted
in Section PS 3150.

Proponents of this view noted that the definition of a tangible capital asset does not specify where the
software has to be physically stored and it does not require the entity to own the host hardware.

View B – The software is expensed in year of acquisition

This view presumes that the inclusion of software in the definition of a tangible capital asset in Section
PS 3150 was not intended to extend to cloud-based software. Proponents of this view indicated that
the footnote in Section PS 3150 was included because at the time it was believed that software was
integral to the use of the computer hardware.

Proponents of this view also indicated that the software is an intangible asset, since it is an asset that
lacks physical substance, and, therefore, is not eligible for recognition in the financial statements.

View C – The software is a non-financial asset; however, it is not a tangible capital asset

This view presumes that software meets the definition of an asset; however, it is scoped out of Section
PS 3150. This view considers the acquisition cost of cloud-based software more in the nature of a
prepaid expense with a limited life – presuming the contract is not in perpetuity and/or without further
charges. This view considers software to be a service and, thus, the initial cost relates to the receipt of
that service over an extended period of time, much like a prepaid multi-year lease.

The Group’s Discussion

Accounting for and understanding cloud computing and its variations is an emerging and significant
issue for standard setters. Group members expressed difficulties in understanding all the permutations
of cloud computing and determining the substance of cloud computing arrangements for accounting
purposes. Although the discussion focused primarily on whether software based in the cloud is a
separate intangible asset or part of a tangible capital asset, a further thought was that the issue goes
beyond considerations of accounting for intangibles. It was noted that the IASB’s discussion also
referenced lease accounting and accounting for property, plant and equipment.

6
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

One view asserted that if the software is customized for one client, then the software is part of a
tangible capital asset even if the software is based on server that hosts the cloud, because the entity
accesses its customized program from its own computers.

Group members agreed that the substance of the cloud computing arrangement would likely affect the
choice of the appropriate accounting treatment. A suggestion was made that if the agreement was for
long-term use, then customization may not be necessary in order to conclude that the software was an
asset of some kind.

Four points were identified for discussion regarding whether they would affect the choice of accounting
treatment (i.e., the location of the software, the definition of an asset and whether it is met for software
based in the cloud, the degree of customization of the software and the acquirer’s control of access to
the software in the cloud).

One Group member noted that if the software was off the shelf (i.e., not customized) and others could
also buy it, then the software should probably be treated like a prepaid expense and be recognized in
expenses over the term of the related agreement.

Another member clarified that even hardware owned by a company needed some software in order to
run (i.e., its operating system such as DOS), and that such software is generally considered to be
integral to the hardware. However, applications such as the Microsoft Office Suite of programs could
be differentiated from software that comprises an operating system – a distinction that Section PS 3150
did not make when it was issued.

Various points were raised about the details and substance of the terms and conditions related to
acquiring and using software based in the cloud:

 What is the window of use of the software by the acquirer – finite, infinite, a prescribed term?

 Who updates the software and does the acquirer have access to updates?

 Can the acquirer take the software back or move it from one cloud provider to another?

 Was the software bought and owned by the acquirer but a third party was asked to host it for use by
the entity?

 Is the arrangement to use the software a license or a service contract or just unrestricted access to
someone else’s asset that is shared with other parties?

The crucial test is the asset definition; whether it is a tangible capital asset or some other non-financial
asset is less important. There was general agreement that such software would comprise some kind of
non-financial asset for the acquirer.

Scenario 2

An entity enters into a shared-cost arrangement where several entities combine resources to jointly
acquire or syndicate multi-year rights to an operating system or enterprise resource planning (ERP)
system software, which is stored in a central location on servers that are not specific to the entity. Each
entity separately contracts with the vendor to either acquire or license the software with its own

7
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

customization requirements. Latecomers may be required to pay for additional media storage to
accommodate their incremental data requirements on the system.

The same three views as discussed for Scenario 1 were considered.

View A – The software is a tangible capital asset

View B – The software is expensed in year of acquisition

View C – The software is a non-financial asset; however, it is not a tangible capital asset

The Group’s Discussion

Initially, the Group’s discussion focused on the nature of the arrangement between the parties to the
ERP system software, whether there was joint ownership and who would report the ERP “entity” in its
financial statements. The submitter clarified that it was akin to an informal joint venture arrangement
but that it truly involved only payment of set-up costs and there was no ERP entity recognized in any
organization’s financial statements. The discussion then moved to how to account for such acquired
software, leaving the consideration of licensing arrangements to Scenario 4.

The discussion essentially recognized that Scenario 2 was not unlike Scenario 1 in both substance and
the Group’s conclusion (i.e., the entity acquires some kind of non-financial asset). Other related issues
raised were accounting for development costs and whether joint control exists. However, the Group
focused on the software question and indicated that each entity that is party to the development of the
software acquired software customized for it and based in the cloud.

Scenario 3

An entity internally develops software that is stored in the cloud.

Generally, the costs of internal development of software are limited to either a dedicated development
function with time and cost allocated to individual projects or costs that are clearly incremental to the
on-going operations of an organization (i.e., contract/consultant costs or internal staff costs that are
back-filled and, therefore, incremental to operations).

The views under Scenario 3 are similar to the views expressed in the Scenario 1. Who developed the
software is not the issue. The issue being analyzed is whether cloud-based software is a tangible
capital asset. Both scenarios include the existence of cloud-based software.

The Group’s Discussion

The Group concluded that this scenario is closest in substance to the circumstances envisioned when
Section PS 3150 was developed. The software is for the use of the one entity that developed it.
Regardless of where it is based, it is customized to work with the other computer hardware and
software of the entity. Thus, through that connection, it is integral to the operations of the entity. The
Group agreed that View A was most appropriate to this scenario.

8
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

Scenario 4

An entity has an ERP system or other software that is licensed for use for a predefined multi-year
period or in perpetuity with an up-front initial payment and annual service, maintenance or upgrade fee.
This software is hosted in a cloud.

View A – Capitalize the initial payment of the software license as a tangible capital asset and expense
the annual fees

This view makes the presumption that in substance, the acquisition of a license to use software is
equivalent to the entity acquiring the software (i.e., that the developer/vendor “sells” the software under
license in order to protect the rights to the code).

This is similar to View A in Scenario 1. Proponents of this view concluded that the inclusion of software
in the definition of a tangible capital asset extends to hosted software. Proponents of this view noted
that the definition of a tangible capital asset does not specify where the software has to be physically
stored and it does not require the entity to own the host hardware. This view takes the position that
there is no difference between software stored on the entity’s own equipment and that of a third party.

This view also presumes the annual fees are maintenance costs as they are incurred to ensure the
software continues to be functional over its useful life.

View B – Capitalize the initial payment and the present value of the future annual payments as tangible
capital assets

This view makes the same presumptions as View A; however, it also presumes that the services
received in exchange for the annual fee are an integral part of the initial acquisition of the software. As
a result, the arrangement is more in the nature of a financing arrangement.

View C – Expense in the year of acquisition

This view presumes that the initial acquisition cost of any software license does not meet the definition
of a tangible capital asset in Section PS 3150.

This view also presumes that the software is an intangible asset and thus not eligible for recognition as
an asset.

View D – Record as other non-financial asset

This view presumes that there is a future economic benefit associated with the acquisition cost of cloud-
based software but that it is not consistent with the intention of Section PS 3150’s inclusion of software
in the definition of a tangible capital asset. This view considers the cost of the license to be more in the
nature of a prepaid expense that would be recognized in expense over the term of the license.
Supporters of this view presume that the license represents future receipt of services – much like a
prepaid multi-year lease, which has a future economic benefit and, therefore, is an asset.

The Group’s Discussion

It was clarified that the ERP system is significantly customized for the entity and that the initial payment
is for that customization.

9
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

Debate centred on the nature of the arrangement (i.e., whether the license was essentially a purchase
of the software as the license was for a long period or for use in perpetuity). If the arrangement is in
substance a purchase, then the software is an asset. And the annual payments for service,
maintenance and upgrades would be an annual expense (View A). On the other hand, one Group
member noted that the maintenance and upgrades might be integral to the functioning of the software;
without such maintenance and upgrades, the software might quickly become obsolete. In such
circumstances, the whole arrangement might be a license, and the potential asset would not be
software but the license.

Another Group member noted that the key is to identify the purpose of the ongoing payments. Are they
a financing component in order to spread cash payments for the software over time, or are they truly
patches and improvements that are integral to the functioning of the software and more in the nature of
betterments to the software? Perhaps crucial to determining the answer is what happens if the annual
payments are no longer made. Can the entity still use the software? In reality, there may be many
variations on access to and the timing of upgrades, as well as the ability to move the software so it can
be maintained and upgraded by another provider.

Various Group members acknowledged that an understanding of the details of the contract and what
their totality indicates about the substance of the arrangement would be the deciding factor for
determining the accounting for the arrangement.

Scenario 5

An entity has an ERP system or other software that is licensed for a fixed term, with no upfront payment
but an annual service, maintenance or upgrade fee. The software is hosted in the cloud.

View A – Capitalize the present value of all minimum future payments and amortize that amount over
the term of the agreement

Supporters of this view would presume the payments to be part of a financing arrangement like a
tangible capital asset lease arrangement, and would argue that in substance, this is no different than a
multi-year licensing agreement that has one up-front payment and annual (lesser) maintenance fees
(as described in Scenario 4), other than the fact that the payments are annualized.

View B – Expense the annual payments as they are incurred

This view sees the payments as an expense of the year for the right to use the software. Proponents of
this view would point to the fact that Public Sector Guideline PSG-2, Leased Tangible Capital Assets,
does not include software in the definition of a leased tangible capital asset and, therefore, does not
consider any assets other than tangible assets to be eligible to record as an asset, notwithstanding the
arrangement may transfer risks and rewards of ownership to the “lessee” in the arrangement.

The Group’s Discussion

As in the other scenarios, Group members felt that the terms and conditions associated with the license
arrangement for software based in the cloud were crucial to determining the appropriate accounting
treatment. It was acknowledged that no Public Sector Accounting Standard or Guideline in the PSA

10
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

Handbook directly addresses this scenario but that looking at its substance and using the GAAP
hierarchy might provide guidance that could be analogized from.

For licenses of one year or less, the choice of View B seemed clear. Even with a longer term, the
license may be just a right to use minimally or uncustomized software based in the cloud that many
other clients also use and the annual payments are just for that use. Again, View B seemed more
appropriate to such arrangements. In contrast, for licenses with longer terms, the guidance in PSG-2
may be useful even if PSG-2 does not specifically include software in the definition of a leased tangible
capital asset. When the terms and conditions are more in the nature of a lease, the software is more
customized and the use of the software is more restricted or exclusive to the entity, View A may be
more appropriate to the substance of the arrangement.

Two Group members asserted that both Scenarios 4 and 5 were really just variations on financing
arrangements for the use of software based in the cloud.

The Group agreed that in all of the scenarios discussed under this issue, the terms and conditions
(responsibilities under the contract including: service, maintenance and upgrades; length of usage term;
degree of access; and, if shared, the degree of customization or uniqueness of the of the software) of
the arrangement to access software based in the cloud are crucial to establishing the substance of the
arrangement for accounting purposes.

The Group indicated that it would be useful for PSAB to issue a standard or guideline, or publish other
tools such as a checklist, in order to address the emerging issues related to accounting for software
based in the cloud.

One Group member noted that there may be some overlap with guidance for intangibles; guidance that
PSAB needs to issue. For example, broadcast rights are also integral to tangible capital assets in that
there needs to be equipment to facilitate broadcasting in order for broadcast rights to be useful.
Further, if such intangibles guidance is issued by PSAB, it might make sense to take software
completely out of the scope of Section PS 3150, and also to differentiate for accounting purposes
between intangibles that are unique to the public sector and those closer in nature to those held in the
private sector.

Another member noted that guidance on accounting for software based in the cloud might be a good
candidate for a Public Sector Guideline. Nuances like those set out in the different scenarios discussed
would benefit from being listed and considered for their relevance for accounting purposes: software on
servers versus owned capital assets; software acquisition versus licensing; paying for development of
software; agreements for service, maintenance and upgrades, etc. A flowchart to help make such
distinctions would be useful in such guidance.

The Group agreed that some update of the standards or guidelines is necessary to take into account
how computing has changed since Section PS 3150 was issued, particularly given the evolution of
cloud computing.

11
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

Sections PS 3260: Liability for Contaminated Sites

Section PS 3260, Liability for Contaminated Sites, indicates that in determining a liability for remediation
of contaminated sites for sites that have not been tested for contamination, consideration should be
given to similar known contaminated sites or sites that have typical or common characteristics. This
submission attempts to determine what constitutes a “typical” or “similar” or “common” site in order to
record a liability estimate when a detailed site investigation has not been completed.

Issue – Should a liability for remediation for a contaminated site be recognized for the
sites that have not been tested for contamination, identified in each of the following
three scenarios?
In all scenarios, it was assumed that an environmental standard existed and that contamination
exceeded that standard, consistent with the requirements of Section PS 3260.

Scenario 1

A site exists where fuel storage tanks were kept in the past (i.e., the fuel tanks are no longer in use).
The entity performs testing on a rotational basis as part of its ongoing monitoring of its sites and testing
to determine the extent of contamination at this site has not been completed. The entity had various
other sites where fuel storage tanks were kept of which approximately 65 per cent of identified sites did
not have contamination that exceeded an environmental standard. The entity has determined that the
most important characteristic that will predict contamination is the type of the tank itself and has
determined that other characteristics (including the location, the geology of the site, the amount of use,
etc.) do not have a bearing on future contamination. The entity historically purchased two types of
tanks. Through an analysis performed on sites that have been tested, the entity has observed that
when Type A tank was used, future contamination would be remote. However, the entity observed that
when Type B tank was used, future contamination is highly likely and Type B tanks represent
substantially all of the 35 per cent of contaminated sites. The site in question had Type B tanks
installed.

View A – A liability should not be recorded since data across all storage tanks indicates that only 35 per
cent of fuel storage tank sites are likely to be contaminated

Under this view, when performing the typical, similar, or common test required by Section PS 3260 to
assess whether information gained at investigated sites can be used to estimate a liability for sites that
have not been investigated, the assessment should be performed across all tank types. To this, the
mere existence of the risk of contamination and some history of contamination does not suggest that all
sites will be contaminated. Given the low probability to date, until the particular site in question has
been tested and contamination confirmed, a liability should not be recorded.

View B – A liability should be recorded upon identification that the site has Type B tanks installed

Proponents of this view indicate that as noted in the scenario, when assessing typical, similar and
common from a recognition perspective, the entity must look to the characteristics that will most likely
result in contamination. Although the historical data does demonstrate there may be some consistency
between the characteristics of historical sites and this particular site (for example, location, geology,

12
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

use, etc.), these other factors have been proven to have no bearing on the contamination and,
therefore, would not be included in the assessment. However, there appears to be a very high
correlation between the factor that has the highest impact on contamination and the actual site in
question. Given that substantially all sites with Type B tanks have contamination and the site in
question has Type B tanks, the entity should record a liability as soon as it is determined that the site
has Type B tanks. For those sites that do not have Type B tanks (i.e., the characteristic that most likely
determines contamination), a liability would only be recorded when the site has been tested and
confirmed to be contaminated.

The Group’s Discussion

A concern was expressed that rigour was needed in making an estimate of a liability for remediation of
a site that is potentially contaminated but has not yet been tested for contamination, because the
estimate might have material financial statement and budget implications.

Nevertheless, the Group agreed that Section PS 3260 requires the accrual of a liability and that testing
of all sites is not needed in order to make an estimate of such a liability. It was noted that there may
need to be reliance on experts such as engineers in estimating the contamination of sites and
assessing the commonality of untested sites with tested sites. Information from tested sites similar to
the untested sites under consideration would be extrapolated to estimate the liability for remediation of
contamination for the untested sites. The Group concluded that using the best and most precise
information available for the estimate is consistent with the application of professional judgment.
Therefore, the results of the analysis that indicated the likelihood of contamination was greater for Type
B tanks, and uncertain for Type A tanks, would support estimating a liability for Type B tanks and not for
Type A tanks. There was recognition that accruals could be adjusted over time as more information
becomes available but that accrual of a liability was both possible and appropriate.

Thus, the Group agreed with View B.

Scenario 2

The entity has lighthouses across its jurisdiction. Many have been painted with lead. Through
repainting exercises, paint flakes have accumulated in the surrounding soil. In certain cases, there have
been cases of mercury and/or petroleum contamination due to the fuel used in operations. Asbestos
has also been used in certain locations.

The location of lighthouses varies. Some are near established communities, some are in remote
locations, and some are on islands with no roadway access. The soil beneath also varies.

Actual costs to remediate the lighthouse locations have varied. Due to the number of lighthouses, not
all have undergone a full site assessment to determine if contamination exists, the extent of
contamination, or plans to remediate.

This scenario also tries to get at how many of the factors in paragraph PS 3260.15 (nature of past
activities on site, location, hydrology, geology, similarities and experience at other known sites) need to
be present to determine the existence of a liability.

13
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

View A – Insufficient information is available to establish a reasonable estimate

Paragraph PS 3260.56 highlights the need for the best estimate to be based on professional judgment
supplemented by experience, third party quotes and, in some cases, reports of independent experts.
Prior to a detailed site examination to confirm contamination and preliminary remediation plans, there is
no liability that can be recognized.

View B – Sites must be similar to be able to apply the results of one site assessment to a second site

Under this view, the types of contaminants, location in relation to relevant aspects (for example, near
populated areas), hydrology and geology must be similar enough that the same remediation plan can
be adopted.

Also, the various aspects that affect the measurement of the liability are assessed individually such that
one can arrive at a base estimate. For example, two sites have the same contamination and are
located near populated areas; however, the geology in one is soil and the other is rock. The entity has
remediated various contaminated sites and knows the costs differentials between rock and soil clean
up. An assessment for the costs of the lighthouse on rock exists. In this case, an estimate can be
attributed to the site on soil. As assessments are completed for the site, the estimate will be revised
over time.

The Group’s Discussion

The discussion initially focused on the practicalities of getting site assessments done, including getting
sufficient funding. The standard has had a big impact on ministry budgets as testing and remediation
costs can be material. It was noted, however, that it was not appropriate to avoid recognizing a liability
by choosing not to test sites. Group members acknowledged the reality that funds can be spent to
clean up the worst sites or to assess lower risk sites and that a balance between the two would be a
policy decision. It was suggested that there is information in the public arena about abatements related
to lead paint and asbestos that would help in the estimation process. In addition, similar to Scenario 1,
there is differential information about the cost of remediating rock versus soil sites that would also aid in
the estimation.

Group members also discussed the need to prioritize the risks for all contaminated sites and allocate
the funds to the highest risk sites. Some indicated that the higher risk sites would also demand the
most effort to estimate the liability for remediation of the contamination given the higher likelihood of
remediation and potential materiality of such sites.

The Group’s consensus on this scenario was that a direct, statistically valid correlation between
assessed and unassessed sites was the best basis for extrapolating the costs from tested sites in
estimating the liability for remediation of untested sites.

Scenario 3

Site A is a coal mine that was in operation for 15 years 50 years ago on which decontamination
activities took place 10 years ago at a cost of $100M. Site B is a coal mine that was in operation for 20
years 20 years ago, on which decontamination activities took place 15 years ago at a cost of $125M.
Site C is a coal mine that was in operation for 50 years 60 years ago and has not been assessed.

14
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

View A – Insufficient information is available to establish a reasonable estimate

Paragraph 3260.56 highlights the need for the best estimate to be based on professional judgment
supplemented by experience, third party quotes and, in some cases, reports of independent experts.
Prior to a detailed site examination of Site C to confirm contamination and preliminary remediation
plans, there is no liability that can be recognized.

View B – The sites are similar as they were all used for the exploitation of coal mines

Under this view, two of the mines had to be decontaminated. Therefore, Site C will also likely need
decontamination. Consequently, a liability is recognized despite the fact that Site C has not been
assessed.

The entity then must assess the measurement of the liability. Options include:

 $112.5M, which represents an average of the two sites;

 $125M because Site B is more similar to Site C in terms of years of operations;

 $312.5M because Site C was in operations for 50 years while Site B was in operations for 20 years
(($125M  20 years) x 50 years); or

 limit the amount to the greatest known amount to remediate a mine (i.e., $125M) to avoid the risk of
over accruing.

The Group’s Discussion

Group members sought a couple of clarifications before discussing the scenario:

 Are the number of years since the mine was in operation or the number of years the mine operated
key characteristics for liability measurement purposes (i.e., with respect to the level of work and /or
cost required to clean up the site)?

 What is the nature of the decontamination required?

The submitter indicated that the level and cost of contamination did not necessarily correlate with either
the number of years since the mine was in operation or the numbers of years the mine was in
operation. The primary determinant of the cost of decontamination is what is found when the mine is
opened for assessment. Thus, the nature of what needs to be done to decontaminate a mine is also
related to the nature of the damage found when the mine is opened for assessment.

The discussion indicated some unease with trying to assess the extent of the liability to be accrued for
remediation of the site given that experience to date has seemed to vary by site. It was noted that
without an assessment by engineers experienced in such decontamination, an estimate of the liability
for remediation of the potential contamination may not be appropriate in this case.

It was noted that many coal mines end up on the government’s hands for decontamination as the
owners have abandoned them and, thus, it would not have been possible for the government to assess
the potential costs of the cleanup when the mines were in operation and accrue a liability for
remediation over time.

15
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

There seemed to be agreement that a liability exists in such a case (View B) but that its measurement
is problematic. Group members indicated that it should be possible to estimate a minimum liability
based on the cleanup costs of other mines. Ideally, it would have been booked over the period the mine
was open as all mines generally require remediation after being closed and often the costs are material.
The estimate of the liability could be refined over time as more information is available.

The Group declined to comment on the measurement options for the decontamination of Site C set out
in View B.

Standard Setting in the Public Interest

PSAB’s Terms of Reference state:

"The Public Sector Accounting Board (PSAB) serves the public interest by establishing
standards and other guidance for financial reporting by all Canadian entities in the public
sector and by contributing to the development of internationally accepted public sector
financial reporting standards. The mission of PSAB shall be to contribute to supporting
informed decision-making and accountability by maintaining a framework that provides a
basis for high-quality information about organizational performance reported by Canadian
public sector entities."

Many have tried to provide a general definition of what is meant by the “public interest”. However,
public interest is an abstract notion, a multi-faceted issue and often defined in terms of the “eye of the
beholder”.

At a previous meeting, the Group discussed what is in the public interest. For example, some Group
members indicated that for standards to be in the public interest, they must result in recognizing and
reporting transactions in an understandable manner. The Group discussed the value of creating a
document to support discussions of why a particular standard is in the public interest.

In response to this discussion and many other discussions on the public interest, PSAB has been
developing a public interest framework of considerations. This framework consists of first order and
second order considerations. First order considerations are those that reflect the conceptual framework
that when present, in and of themselves, provide significant assurance that the public interest is being
served. Second order considerations are those that may arise throughout the due process that may
also be considered. While they contribute to determining whether the public interest is being served, in
and of themselves, any one consideration does not significantly contribute to the public interest.

PSAB’s Chair, Rod Monette, introduced the session, indicating that PSAB already looks at issues
through a public interest lens but that a framework of considerations would make that practice more
transparent and consistent. Mr. Monette gave some examples of issues where accounting may not be
understandable to the public, noting that PSAB feels that it is important when looking at different
technical options to consider their effect on the public interest. Mr. Monette indicated that PSAB is not
proposing a definition of the public interest, because what serves the public may mean different things
to different people at different times. Instead, PSAB is looking at developing considerations that will
guide PSAB in its deliberations rather than a definition of the public interest. In illustrating the role of
the conceptual framework in PSAB’s deliberations, Mr. Monette also gave an example of the balance

16
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

that may need to be struck between qualitative characteristics such as comparability and
understandability, indicating that such trade-offs may need to have visibility – both in standard setting
and in preparers’ decisions about accounting policies and financial reporting. Mr. Monette highlighted
some of the changes made to PSAB’s draft public interest framework since the Group discussed it in
May 2015.

The Group’s Discussion

1
Group members sought clarifications – in particular, the purpose of PSAB’s public interest framework.
Mr. Monette responded that the purposes would be:

 to promote thinking about the public interest in the course of standard setting as the journey may be
as important as the destination;

 to consider publishing PSAB’s public interest considerations so that its constituents would
understand them;

 to keep the public interest top of mind over the course of the revisions to the conceptual framework;
and

 to frame PSAB’s strategic discussions with its oversight body, the Accounting Standards Oversight
Committee.

Group members indicated that the considerations were a positive step for PSAB. One Group member
asked that the considerations be linked to PSAB’s Mission in the public interest framework itself.

Some concern was raised about how to ensure that the considerations were applied consistently. One
Group member indicated that it should be made clear in the document that it was good to check for
consistency in decisions and standards with the work of other standard setters but that there might be
good theoretical and/or practical reasons why the Canadian public sector approach should be
consciously different.

The Group considered what would be the fair expectation of the public’s level of understanding of public
sector financial reporting, indicating that there was a balance to be struck between making financial
reports digestible by the broader public and the need to ensure that complex information is not
excluded from financial statements if it is needed to provide the full financial picture to the public.

Another Group member noted that the public wants to understand a government’s fiscal policies to the
extent that they will affect their taxes and level and type of services. It was noted that financial
statements between governments might not be comparable because of differences in the underlying
fiscal frameworks, which may affect the economic substance of what is done by individual jurisdictions.
Thus, comparability may not be desirable or achievable.

A few Group members made it clear that the conceptual framework is the touchstone for applying the
first and second order considerations being considered by PSAB. The presumption has to be that the

1
The public interest framework is a document for use by PSAB in the standard setting process. It does not form part of the
conceptual framework, which is integral to the PSA Handbook. Proposed revisions to the conceptual framework will undergo
scrutiny through PSAB’s formal due process.

17
 Report on Public Meeting on October 27, 2015 – Non-authoritative material

conceptual framework is developed as the test as to what is in the public interest and that the second
level considerations are of less importance. It was noted that the first order considerations, as currently
proposed, essentially consist of the qualitative characteristics of financial information that are in the
existing conceptual framework in the PSA Handbook. And debates over the right accounting treatment
will often come down to the different weights put on these characteristics, including those that try to
balance comparability and understandability.

It was indicated that clarity and consistency within the conceptual framework and standards would allow
for their easier application. Solid accounting standards with rigour reduce the moral hazard of
misinformation in financial reports and should reduce the gap between the public’s understanding of
how they are taxed and how their taxes are used. It is important that the public view the financial
reports produced in accordance with the standards as being trustworthy because the standards are
seen to promote the production of trustworthy information for the public.

The Group discussed the merits of publishing the public interest framework when it is finalized and
suggested how it could be communicated.

One Group member commented that some feel that accounting standards do not reflect economic
reality but try instead to shape it. In fact, accounting standards try to portray economic reality in as
neutral a fashion as possible. Another Group member reflected that comparability across jurisdictions
may not be possible because the organizations included in the various government reporting entities
might be different as a result of the structural and accountability relationships within each such entity. It
was noted that comparability cannot be forced. For example, universities should not all be consolidated
by their respective provincial governments unless each province controls its universities.

One Group member contrasted comparability and consistency, noting that they are different. The
member also noted that the public interest framework is a useful document that leverages the
conceptual framework and that the revised conceptual framework should make its way into the public
interest framework when it is finalized rather than the public interest framework being forced into the
conceptual framework.

***************

The Group’s Chair, Andrew Newman, concluded the meeting by mentioning the retirement of PSAB
Director, Tim Beauchamp, and thanking Tim for his contribution to public sector accounting and serving
the public interest. He thanked Tim for his commitment to and resourcing of the Group. And he
commented that there would be continued application of Tim’s work with PSAB for many future years.

18